Development

Every team reinvents Git workflows. Most end up with something that worked for three people but breaks at fifteen. Here’s what actually scales. The Problem With “Whatever Works” Small teams can get away with anything. Push to main, YOLO merges, commit messages like “fix stuff” — it all works when you can shout across the room. Then the team grows. Suddenly: Two people edit the same file and spend an hour on merge conflicts Nobody knows what’s in production vs staging “Which commit broke this?” becomes an archaeological dig Releases are terrifying because nobody’s sure what changed The solution isn’t more process. It’s the right process. ...

Feature flags turn deployment into a two-step process: ship the code, then enable the feature. This separation is powerful when done right and a maintenance nightmare when done wrong. The Core Value Proposition Without feature flags, deployment equals release. Ship broken code? Users see it immediately. Need to roll back? Redeploy the previous version. Want to test with 1% of users? Build custom infrastructure. With feature flags, you decouple these concerns: ...

Bad APIs create support tickets. Good APIs create fans. Here’s how to design APIs that developers will actually enjoy using. The Fundamentals Use Nouns, Not Verbs # P G P # P G D O E O O E E B S T S G S T L a T T o T E d o T g / d u E c e d u s r t e s e / e U l e r u a s e r s s t e t s e e r e r U s U s s s / e e 1 r r 2 / 3 1 2 3 The HTTP method IS the verb. The URL is the noun. ...

Everyone has access to AI coding assistants now. Most people use them poorly. Here’s how to actually get value from them. The Mental Model Shift Stop thinking of AI assistants as “autocomplete on steroids.” Think of them as a junior developer who: Has read every Stack Overflow answer ever written Types infinitely fast Never gets tired or annoyed Has no memory of what you discussed 5 minutes ago Will confidently produce plausible-looking nonsense That last point is crucial. These tools don’t know things. They predict likely tokens. The output often looks right even when it’s wrong. ...

Docker Compose turns “works on my machine” into “works everywhere.” Here’s how to structure it for real development workflows. Basic Structure 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 # docker-compose.yml services: app: build: . ports: - "3000:3000" volumes: - .:/app environment: - NODE_ENV=development db: image: postgres:15 environment: POSTGRES_PASSWORD: devpass Start everything: ...

Every Python project should have its own virtual environment. It’s not optional — it’s how you avoid dependency hell, reproducibility issues, and the dreaded “but it works on my machine.” Why Virtual Environments? Without virtual environments: Project A needs requests==2.25 Project B needs requests==2.31 Both use system Python One project breaks With virtual environments: Each project has isolated dependencies Different Python versions per project Reproducible across machines No sudo required for installing packages The Built-in Way: venv Python 3.3+ includes venv: ...

Docker Compose started as a development tool but has grown into something usable for small production deployments. Here are patterns that make it work well in both contexts. Environment-Specific Overrides Base configuration with environment-specific overrides: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 # docker-compose.yml (base) services: app: image: myapp:${VERSION:-latest} environment: - DATABASE_URL depends_on: - db db: image: postgres:15 volumes: - postgres_data:/var/lib/postgresql/data volumes: postgres_data: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 # docker-compose.override.yml (auto-loaded in dev) services: app: build: . volumes: - .:/app - /app/node_modules environment: - DEBUG=true ports: - "3000:3000" db: ports: - "5432:5432" 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 # docker-compose.prod.yml services: app: restart: always deploy: replicas: 2 resources: limits: cpus: '0.5' memory: 512M logging: driver: json-file options: max-size: "10m" max-file: "3" db: restart: always Usage: ...

Code review is a skill separate from coding. A great programmer can give terrible reviews — nitpicking style while missing logic bugs, or rubber-stamping everything to avoid conflict. Good reviews improve code quality and team knowledge. Bad reviews slow everything down. These practices help you do reviews that actually help. The Reviewer’s Mindset You’re Not the Gatekeeper ❌ ✅ " " I D o w e o s u l t d h n i ' s t w h o a r v k e c w o r r i r t e t c e t n l y i t a n t d h i m s a i w n a t y a " i n a b l y ? " Your job isn’t to make the code match your personal style. It’s to catch problems and improve clarity. ...

A well-designed API client turns complex HTTP interactions into simple method calls. It handles authentication, retries, errors, and serialization — so users don’t have to. These patterns create clients that developers actually enjoy using. Basic Structure 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 import httpx from typing import Optional from dataclasses import dataclass @dataclass class APIConfig: base_url: str api_key: str timeout: float = 30.0 max_retries: int = 3 class APIClient: def __init__(self, config: APIConfig): self.config = config self._client = httpx.Client( base_url=config.base_url, timeout=config.timeout, headers={"Authorization": f"Bearer {config.api_key}"} ) def _request(self, method: str, path: str, **kwargs) -> dict: response = self._client.request(method, path, **kwargs) response.raise_for_status() return response.json() def close(self): self._client.close() def __enter__(self): return self def __exit__(self, *args): self.close() Resource-Based Design Organize by resource, not by HTTP method: ...

Everyone agrees testing is important. Few agree on how much, what kind, and where to focus. The testing pyramid provides a framework: many fast unit tests, fewer integration tests, even fewer end-to-end tests. But the pyramid only works if you understand why each layer exists. The Pyramid E I U 2 n n E t i e t T g e r T s a e t t s s i t o s ( n f ( e T m w e a , s n t y s s , l o ( f w s a , o s m t e e , x , p c e m h n e e s d a i i p v u ) e m ) s p e e d ) Unit tests: Test individual functions/classes in isolation Integration tests: Test components working together E2E tests: Test the full system like a user would ...